Placebo! Nocebo!

April 17th, 2010 by Ben Goldacre in onanism, placebo, podcast | 29 Comments »

Hi, two quick videos… one on the mighty placebo effect, via NHS Choices, one on the nocebo effect from Nerdstock.

Sorry no column today, something pretty urgent came up on Friday, back next week, there’s this on libel from yesterday’s paper, and I’ll see if I can hustle what the column would have been into the paper on Mon/Tues.

If you like what I do, and you want me to do more, you can: buy my books Bad Science and Bad Pharma, give them to your friends, put them on your reading list, employ me to do a talk, or tweet this article to your friends. Thanks! ++++++++++++++++++++++++++++++++++++++++++

29 Responses

  1. bt42 said,

    April 17, 2010 at 3:04 pm

    Hi Ben, can you please direct me to the article for that last study you mentioned in the video, with the muscle relaxant? Cheers!

  2. ignoranceisalearnedbehavior said,

    April 17, 2010 at 4:18 pm

    So advertising Meds on TV, Radio, in newspapers etc, as they do in the US, along with long lists of side-effects might be rather counter-productive then. Ooops.

  3. marpla said,

    April 17, 2010 at 4:35 pm

    What is then the relationship between homeopathy and placebo-effekt?

  4. Jerry said,

    April 17, 2010 at 5:31 pm

    hopeopathy relies on the placebo effect for it’s anectdotal success stories.

    Unfortunately the placebo effect isn’t reliable.

  5. caitlinburke said,

    April 17, 2010 at 5:40 pm

    For the carisoprodol study: ?

  6. caitlinburke said,

    April 17, 2010 at 5:44 pm


  7. iamjohn said,

    April 17, 2010 at 5:53 pm

    I didn’t know you were funny

  8. conejo said,

    April 17, 2010 at 8:55 pm


    I posted this in Liberal voice, please forgive the copy but I said what I wanted to say and it doesn’t really need editing.

    ” … “mind over health”. In fact there is already a well established method for exploiting the phenomenon: hypnosis/hypnotherapy. Now don’t get me wrong: I am not here to proclaim that it is ‘the answer’ or that it is anywhere near to being well enough developed. However I would like to make a case for more consideration. A few years ago I trained and qualified as a hypnotherapist, but have never practiced. So I know at least a little of what I am talking about but do not have a vested interest.

    First, and above all, get out of your heads the nonsense of stage hypnosis which is conjuring with minds for entertainment. Surgery may still take place in a ‘theatre’ but is no longer an amusement: I hope hypnosis will go through a similar transition. Also abandon any idea that therapists can ‘program’ victims: neither to rob banks, commit murder nor heal themselves. That’s the stuff of Hollywood. A fiction.

    Hypnotherapy probably works by exploiting the placebo (or similar) effect – but it is *honest* about doing so, no lying is necessary. Even the language used is clear: it works by the power of ’suggestion’; it works by exploiting imagination and fantasy; the therapist enables and facilitates – s/he is not like a mechanic ‘fixing’ a problem with a car. Certainly the therapist is positive and assertive about the value of their treatment – but then so is my GP. In fact saying “this course of hypnotherapy is going to work because it will exploit your placebo response” would probably be regarded by most hypnotherapists as a useful way of increasing the confidence of the client.

    Hypnotherapy is already used by some conventionally trained doctors and dentists (although I was shocked to discover how superficial their training was compared to the longer program I went through, which itself was described by the tutor as only a beginning). There is some uncertainty about a systematic explanation of how it works, but that revolves around a lack of understanding about ‘the sub-conscious’, the mind, and even about an agreed set of concepts and words that can be used to discuss the topic. However – importantly – it does not posit an incredible mechanism (as homeopathy does) and it does lend itself to scientific trials. My experience of the people who were involved in hypnotherapy is that they were open to a rigorous scientific examination of the subject without any of the special pleading that most ‘magic’ medicine seems to wheel out: none of the ‘our method is special and can’t be tested by impersonal scientific studies’.

    Does it work with ‘everything’? Certainly not. Does it work with ’some’ things? I believe it does and I believe it could work with many more disorders and could certainly assist with many more issues around chronic disease such as pain management. Does it need more development, more research and more evidence? Definitely. Will it get it? Ah, there’s the rub: probably not because it is cheap. People can be relatively easily trained and are not highly paid (compared to GPs, surgeons, etc). Business is not going to make enough profit out of it to justify spending the money to carry out large-scale trials. Pity.”

  9. eleanor said,

    April 17, 2010 at 9:17 pm

    Of course it could be that rather than being dramatic or serious, the injection stimulates pain receptors. And the colour of the pills could be related to associations with the effects of blue light.

  10. geodoc said,

    April 17, 2010 at 9:24 pm

    Good stuff.

    The whole point is that the placebo effect is reliable, and incredibly and unexpectedly powerful. Of course there are certain situations it can’t work in (won’t reduce a fractured leg etc), but there are many others in which it will reliably work and these include, but fascinatingly are not limited too, those with an obvious psychological component. Conventional medicine may struggle with these sorts of conditions: chronic fatigue syndrome, chronic pain etc, so there’s a real opening for homeopathy and other quackery to come in and make use of the phenomenon.

    While I agree with everything Ben says in his first video, there’s a kind of naive optimism there- it’s great to say let’s use it to improve treatments we already know work, but those treatments have got to be there in the first place. The strength of the placebo effect is such that both doctors and patients will implicitly collude to use it, particularly in situations where previous treatments have been tried without success. I’m sorry Ben, but it’s inevitable. When this happens I’d prefer the placebo to be a sugar pill (or simple linctus) rather than, say, amoxicillin.

    In practice, it’s handy to have a few safe, cheap but probably non-beneficial (beyond placebo effect) treatments handy which haven’t yet been completely consigned to oblivion by evidence-based medicine, which doctors can recommend saying ‘many people seem to find this helpful [which will be true], why not give it a try?’. Thus the doctor’s conscience is salved— there’s still the possibility of a non-placebo benefit— while the power of the placebo effect is deployed, perhaps in a situation where other treatments haven’t helped, or even been harmful.

    Glucosamine, anyone?

  11. Scented Nectar said,

    April 17, 2010 at 9:35 pm

    That’s some interesting stuff to think about. Before, I would have said in a blanket statement that all placebos are immoral, harmful to more than are helped, etc. I never considered the nonharmful, *moral* use of them (as in packaging colour psychology, not something like telling them they are getting more of the drug than they really are) in combo with very real drugs in order that they may be of additional benefit to anyone whose treatment may actually be enhanced by the placebo effect.

  12. amaca said,

    April 17, 2010 at 11:26 pm

    How about some evidence based non-medicine?

    According to this Cochrane Institute report, repeated three times and with some verification:

    “Placebo interventions are often claimed to substantially improve many clinical conditions. However, most reports on effects of placebos are based on unreliable studies that have not randomised patients to placebo or no treatment.
    “We studied the effect of placebo treatments by reviewing 202 trials comparing placebo treatment with no treatment covering 60 healthcare problems. In general, placebo treatments produced no major health benefits, although on average they had a modest effect on outcomes reported by patients, such as pain. However, the effect on pain varied from large to non-existent, even in well-conducted trials. Variations in the effect of placebo was partly explained by variations in how trials were conducted, the type of placebo used, and whether patients were informed that the trial involved placebo. ”

    Hróbjartsson A, Gøtzsche PC. Placebo interventions for all clinical conditions. Cochrane Database of Systematic Reviews 2010, Issue 1. Art. No.: CD003974. DOI: 10.1002/14651858.CD003974.pub3.

  13. micktransit said,

    April 17, 2010 at 11:35 pm

    Although the placebo response is better known, the nocebo response is actually easier to understand.
    I suspect that in the modern world, with information overload and constant health scares, we’re all suffering a constant background nocebo effect. What appears to be a placebo effect may be in fact just the constant nocebo effect being temporarily relieved. This sounds unnecessarily complicated, I know, but it does make more sense to me.
    Furthermore, whatever real harm may be done by radioactivity, mobile phones, passive smoking, etc; must inevitably be compounded by publicity fuelling our expectations of harm. Okay, you can laugh that off if you want, but it makes sense. There’s an ethical dilemma there. A serious one. This is not trivial.

  14. iliff said,

    April 18, 2010 at 12:06 am

    The YouTube auto-captions are hilarious.

    “… yeah more confident people and I can in ninety six and ten the masters personal that’s all this wishy-washy but really genuinely incurring the outcomes of all treatments and I think that’s the real key if the hearings what the to see but if a person’s tie this crime to find way of taking treatments which we or deny with genuine effective and make them even back for more information the state’s WWW dot any chance don’t UK”

  15. mikewhit said,

    April 18, 2010 at 9:57 am

    we’re all suffering a constant background nocebo effect

    Only if we read the Daily Mail …

  16. RogerMexico said,

    April 18, 2010 at 1:02 pm

    I’m going to stick my head over the parapet here, but surely there’s a point here that, with the way it’s done on the NHS in Britain, homeopathy is almost the perfect way to manage and maximise the placebo effect.

    Patients are referred to conventionally trained doctors who can and should spot anything serious and get it treated properly (and can be jumped on by the GMC if not). Water is about as harmless as you can get. Side effects, though possible tend to be a lot less unlike say antidepressants. And it’s got to be cheaper than throwing pharmaceuticals down people’s throats.

    Like it or not doctors have to deal with a lot of situations where there’s not much they can do: self-limiting conditions, the worried well, vague aches and pains investigated to death, conditions best left alone despite patient demands. If they can say “Go and see Dr Woo the Homeopath – he’s had some success with problems like these”, at worst they’re getting a medical second opinion. At best the placebo effect works its magic charm and the patient feels better.

    It may disturb some people’s sense of medical ethics, but given that the most basic ethical rule is “do no harm”, I can’t see how the current use of homeopathy on the NHS fails that.

  17. HypnoSynthesis said,

    April 19, 2010 at 12:02 am

    Blimey! When are you going to become a hypnotherapist???

    Do you realise that there’s about 100 years of research on suggestion in medicine in the field of hypnotism? Medline has nearly 10,000 citations indexed under the heading of hypnosis. The problems hypnosis is most effective in treating happen to be the ones most responsive to placebos, and the same is true for psychotherapy in general.

    James Braid, the Scottish skeptic and surgeon who invented hypnotism (and whose collected works I edited!) defined it as “focused attention on a dominant expectant idea”, i.e., concentrated autosuggestion.

    Prof. Irvine Kirsch at Hull is one of the leading contemporary hypnosis researchers and has also developed a cognitive psychological theory of the placebo response (“response set” theory). He defined hypnosis as a “non-deceptive mega-placebo”, which is like a whole theory compressed into two words.

  18. skyesteve said,

    April 19, 2010 at 9:10 am

    @RogerMexico – sorry, but in my book “hoodwinking” someone and not telling them that that is what you are doing is little more than deception and I regard deception as “harmful”. And recommending a treatment which you yourself believe to be junk is downright dishonest unless you tell your patient that you believe that it’s junk. The scenario you describe, in my view, merely serves to perptuate unnecessary demand.
    There is a spurious argument that homoeopathy does no harm but it does if it creates unnecessary dependancy, perpetuates non-disease or causes patients not to seek established, effective evidence-based therapy for a potentially serious condition such as asthma.

  19. AJH said,

    April 19, 2010 at 6:47 pm

    Did Ben say “8 million metric fucktonnes?” That’s genius right there.

  20. RogerMexico said,

    April 19, 2010 at 7:18 pm

    @Skyesteve – but you’re neither deceiving or lying to them. A

    non-homeopathic doctor referring a patient to a homeopath will only send her/him to a doctor who has had success with these treatments (due to placebo – but we all know how powerful that is). And the

    homeopathic doctor her/himself believes in the treatments (with

    added efficacy – that’s why trials are double blind).

    And surely what matters to the patient is not whether a treatment

    is “junk” but whether it works. It’s true we don’t understand the

    mechanism of the placebo effect, but should medicine then refuse to

    use any treatments or drugs where it doesn’t understand completely

    how they work. Logically, as no one would say there is complete

    knowledge of how the body and mind (co)operate, doctors should

    therefore refuse any treatment whatsoever in any circumstance.

    It’s interesting that you complain about perpetuating “unnecessary

    demand” and “non-disease”. I don’t think that people always

    automatically feel completely well once you tell them there’s

    nothing wrong with them. Aren’t they then more likely to wander

    off to seek treatment in the wild and wacky world of “alternative

    medicine”. Remember that I’m advocating NHS homeopathy where the

    practitioners are also conventionally trained and competent.

    Of course you’re also assuming that no doctor will ever make a

    misdiagnosis or miss a condition and all things that can go wrong

    with the human body have now been discovered.

    In the example of asthma, you raise an interesting case. Not of

    vital treatment being missed (it’s NHS qualified doctors still)

    but of over-diagnosis and over-treatment. With some conditions it may be that passing symptoms are being diagnosed a chronic conditions – in part because it reassures patients to “know what they’ve got”. There’s also the question of whether certain conditions should automatically be aggressively treated (prostate cancer comes to mind) or whether monitoring first is better.

    If you want, you can adopt the interventionist model, defensively throwing every possible treatment at patients. But this usually has the interests of the medical insurers in mind rather than the patients’ quality of life. Alternatively you can argue that something such a homeopathy can reconcile worried patients to a situation when benign neglect (plus monitoring if appropriate) is the best, if less dramatic, form of action.

    As already has been pointed out, doctors hand out effective placebos all the time: antibiotics for viruses; anti-depressants for mild depression (now there’s an “unnecessary dependency”). What NHS homeopathy does is maximise the placebo benefit for patients within a safe environment.

  21. RogerMexico said,

    April 19, 2010 at 7:23 pm

    Sorry about that – program took formatting into its own hands. Imagine your own paragraphs.

  22. MedsVsTherapy said,

    April 19, 2010 at 7:41 pm

    With my slim training and experience in hypnotherapy, I agree with Cojones. My efforts, on a chronic pain treatment training rotation, were to achieve an intellectual conceptualization of what the heck happens, while figuring that it would take a lot of clinical trial-and-error to actually reduce pain for a pain patient. -I imagine it comes with practice, like a golf swing. I was remarkably successful with one of a few pain patients, and my explanation is that I was kind of on-target and he believed in me (two clocks will each be wrong twice a day, and if the clocks are both set to the same time, then…). It is odd to be thoroughly honest about attempting chronic pain reduction through suggestion and some imagination. Odder to have a pt. exclaim great relief, gazing at you as if you were Penn and or Teller.

    For me, the value of stage hypnotism – the comedy show hypnotists – is to grasp the power of hypnotism. Youtube has these, but to grasp it, you must attend a show in person. Only then can you see convincingly real people queuing up, paying their cover charge, getting screened for hypnotizability, getting induced (or whatever you call it), responding, etc., and after the show ends, you are able to hear multiple groups of show patrons discuss with their recently-hypnotized friend what he or she has done in the past half hour.

    Well worth the money to any amateur psychologists, professional psychologists, and to anyone concerned with how placebo effects might work. Cluck.

  23. amaca said,

    April 19, 2010 at 8:20 pm


    A GP can’t refer patients to a homeopathic doctor who has had success with these treatments because there’s no way of telling whether success was due to placebo or whether success was due to any condition being naturally cured (or regression to the mean, or misdiagnosis, or spontaneous recovery, or whatever).

    Nor, strictly speaking, should a doctor refer someone to a homeopath of any sort since to do so means referring to someone who is in breach of their own ethical framework (lying to a patient, selling snakeoil for profit, that sort of thing).

    Nor, according to the Cochrane Institute, is the placebo effect capable of any sort of cure anyway. All it can help with is secondary symptoms and they found no evidence that placebo has any effect on anything without a psychosomatic element. Placebo is good for pain, but if a doctor has worked their way up from paracetomol through morphine and the more serious opiates, do you think it’s likely that sugar water is really going to do the trick?

    What NHS homeopathy does is lie to the patient, encourage them to use magic rather than evidenced medicine (and next time, they might just go and homeopathic sugar from Boots when they’re actually ill) and encourage them to believe that taking pills is the way to mend anything; what is the placebo effect of a doctor with a good bedside manner telling a patient that their condition will heal in time and how to minimize the discomfort in the meantime?

  24. pajamapaati said,

    April 20, 2010 at 12:05 am


    >According to this Cochrane Institute report, repeated three times and with some verification:

    Here’s a link:

    Ben: I think you really need to address this. Your book and this site strongly presuppose that there is a real and significant placebo effect (the videos above refer to placebo effects on gastric ulcers and muscle relaxation), but also tell us that Cochrane reports are the Gold Standard for evaluation of medical evidence — and here’s Cochrane dissing placebo!

  25. skyesteve said,

    April 20, 2010 at 8:33 am

    Hi RogerMexico – thank you for your reply. I see where you are coming from it’s just that I don’t agree. In my view all the best relationships are built on openess and honesty. I will refer someone to an NHS homoeopath if they want me to – but not before I have told them that I know of no evidence that it’s any better than placebo and that I do not think it should be funded by the NHS. But that’s just my view.
    On the more specific point you make:

    If you want, you can adopt the interventionist model, defensively throwing every possible treatment at patients. But this usually has the interests of the medical insurers in mind rather than the patients’ quality of life.

    I don’t think this reflects the kind of general practice I practice where “wait and see” is a day-in-day out part of the job and where insurers have no bearing on my work as I have no private patients.

  26. janewhitman said,

    April 26, 2010 at 2:26 am

    the article i read was here…

  27. pajamapaati said,

    April 27, 2010 at 12:07 am

    is SPAM!

  28. mikeatanginadotorg said,

    May 4, 2010 at 5:09 pm

    For those of us lucky enough to have the freedom to practice patient-centred medicine, placebo comes with the package. Why wouldn’t you want to maximise the effects of an intervention when it is safe and ethical to do so? The ethics of exploiting the placebo effect is crucial and Ben rightly says that lying to patients is unethical, and suggests that doctors don’t lie about treatments. But what about encouraging patients to believe something that isn’t true by witholding facts? Isn’t that also unethical?

    Take angioplasty for stable angina for example. Many patients think it is necessary to prevent heart attacks and most assume that it has been subject to rigorous scientific evaluation. They would be wrong on both counts. Outside a heart attack situation, angioplasty is costly, does not improve life expectancy, exposes the patient to risk of serious harm and has never been compared to a sham procedure. The two trials of sham angina surgery, performed fifty years ago, both demonstrated that sham surgery was a powerful placebo. Consequently, the small and temporary benefits observed in clinical trials of angioplasty could easily be explained by a placebo effect.

    Ben has rightly challenged medical quackery. Homeopathy was a fair and ‘easy’ target, but isn’t it time to ask whether modern snake oil salesmen have managed to successfully conceal themselves in masks and theatre blues? Just because palliative angioplasty produces obvious and measurable physical changes in arteries, it shouldn’t be let of the evidence-based hook. About £1billion has been put aside over the next four years to fund scheduled angioplasty. A lot of that will be spent on foreign imports to achieve something that might just be a very expensive and potentially lethal placebo.

  29. Astonmartin said,

    January 2, 2011 at 4:55 pm

    I believe in the power of mind over matter (the placebo effect), ive done alot of background reading on it, I found it very interesting and its obvious there are countless evidence to say it work especailly in areas of Anxiety and depression which is clearly psychological.

    With me looking at the whole health industry on a business point of view would it be a worthwile investment to take on teachers who are train Psychologists and teach these youngsters in school, at an early age, of the power of placebo to cure their illnesses and how to implement it. I know there are teachers teaching psychology in senior schools but do they teach the kids about the power of thought?

    The long term benefits, I believe, there will be less people running to their Doctor and the HNS can save millions of £s and they concentrate on more serious matters such as trauma and accidents.